Magnesiation of Alkyl Fluorides Catalyzed by Rhodium–Aluminum Bimetallic Complexes

Abstract

Since the pioneering work by Grignard in 1900, organomagnesium reagents, the so-called Grignard reagents, have been indispensable in organic synthesis. Alkyl Grignard reagents are usually prepared from the corresponding alkyl iodides, bromides, or chlorides with Mg, whereas alkyl fluorides are not viable substrates under conventional conditions due to the high stability of the C–F bonds. We report that Al–Rh bimetallic complexes catalyze the magnesiation of C(sp³)–F bonds of alkyl fluorides using easy-to-handle Mg powder. The present conditions can accommodate primary, secondary, or tertiary alkyl fluorides to afford the corresponding alkylmagnesium reagents, which can be successfully converted into various functionalities.

Since the seminal discovery of organomagnesium reagents (Grignard reagents) by Victor Grignard,[2a] they have been indispensable in organic synthesis due to their exceptional reactivity toward various electrophiles, facilitating the construction of a wide variety of organic molecules.[2b] Alkylmagnesium reagents have conventionally been synthesized from alkyl halides (R–X; X = Cl, Br, I) and Mg turnings or powder (Scheme [1]A).[2b] Knochel and co-workers have pioneered the halogen–magnesium exchange reaction of alkyl halides with i-PrMgCl∙LiCl (turbo Grignard reagent) as another reliable and practical method for preparing Grignard reagents, enabling the preparation of functionalized alkylmagnesiums.[3]

However, the preparation of alkylmagnesium reagents from alkyl fluorides has been difficult due to the high bond-dissociation energy of C–F bonds (CH₃–X; X = F: 110 kcal/mol vs X = Br: 70 kcal/mol, X = H: 105 kcal/mol).[4] To magnesiate the C–F bond of alkyl fluorides, harsh reaction conditions (high temperatures and prolonged reaction times)[5] or pyrophoric highly dispersed magnesium reagents are required (Scheme [1]B).[6] For example, the magnesiation of 1-fluorohexane proceeds with only moderate efficiency, even with a large excess of Rieke Mg in THF.[6a] These protocols permit the magnesiation of primary alkyl fluorides, whereas magnesiations of secondary and tertiary ones remain elusive. Recently, Crimmin and co-workers demonstrated a magnesiation of primary, secondary, and tertiary alkyl fluorides by using β-diketiminate-stabilized Mg(I)–Mg(I) complexes, prepared by using stoichiometric amounts of alkaline metals.[7] There have been no reports of general methods for the magnesiation of alkyl fluorides with readily available and easy-to-handle Mg agents.

We recently developed a magnesiation of aryl fluorides catalyzed by the Al–Rh bimetallic complex 1a, in which the Ar–F bonds were effectively activated by the polarized Al–Rh bond in a cooperative manner (Scheme [2]A).[8] Theoretical calculations and stoichiometric experiments have revealed that the cooperative Ar–F bond activation requires a small activation barrier (3.7 kcal/mol). Accordingly, we expected that the Al–Rh bimetallic complexes might also be effective in the catalytic magnesiation of alkyl fluorides through C(sp³)–F bond activation in a similar manner (Scheme [2]B).

The catalytic C(sp³)–F bond magnesiation of 2-(3-fluorobutyl)naphthalene (2a; 1.0 equiv) with Mg powder (5.0 equiv) was conducted in the presence of the bimetallic Rh complexes 1a and 1b in THF at 80 °C for 47–48 hours (Scheme [3], top). 2-Methyl-4-(2-naphthyl)butanoic acid (3a) was obtained in 71% NMR yield with 1a (5.0 mol% of Rh) as a catalyst after quenching the reaction mixture with CO₂ (1 atm) followed by an acidic workup with 3 M aq HCl. 2-Methyl-4-phenylbutanoic acid (3b) was similarly obtained in a high yield, even in the presence of a reduced amount (1.0 mol%) of the catalyst. Al–Rh complex 1b, bearing isopropyl groups on the phosphorus atoms instead of phenyl groups, also produced 3a in 63% NMR yield. The reaction was completely suppressed in the absence of the Al–Rh complexes 1. Combinations of [RhCl(nbd)]₂ (5.0 mol% of Rh), ligand (10 mol% of L), and Et₂AlCl (20 mol%) did not convert 3-fluoro-1-phenylbutane (2b) (Scheme [3], bottom), whereas the [RhCl(nbd)]₂/PPh₃ system without Et₂AlCl interestingly gave 3b in 62% NMR yield (Al–Rh complex 1a afforded 3b from 2b in 73% yield; see Scheme [4]). Other phosphine ligands, such as DPEphos, also afforded 3b in moderate yield (42–62%). Note that 1a showed a higher reaction rate to yield 3b than did the Rh/PNNNP(L) or PPh₃ system, indicating that the Al ligand plays a key role in this reaction. Accordingly, Al–Rh complexes 1 were found to be the most effective for the catalytic magnesiation of alkyl fluorides. The products were found to be the corresponding dialkylmagnesium species (possibly produced through the formation of insoluble MgF₂),[9] based on NMR analyses of the reaction mixture before quenching with electrophiles [for details, see the Supporting Inform<ation (SI), Figures S2–S5].

Various fluoroalkanes were subjected to the optimized conditions with catalyst 1a on a 0.50 mmol scale (Scheme [4]). The reaction using a primary alkyl fluoride, octyl fluoride (2c), afforded octane (3c) in 72% GC yield after quenching of the reaction with H₂O. The magnesiation of the secondary alkyl fluoride 2b, followed by reactions with a variety of electrophiles [CO₂, H₂O, i-PrOB(pin), the Weinreb amide,[10] or O₂] afforded the corresponding products 3b and 3d–3g in good to moderate yields. Notably, regioisomers and/or products derived from β-hydride elimination were not observed. β-Hydride elimination could be slow due to the sterically demanding and rigid PAlP ligand in 1a, which could block a vacant coordination site.[11] Fluorocyclohexane (2d) was a viable substrate, and even the tertiary alkyl fluoride 1-fluoroadamantane (2e) gave adamantane (3i) and 1-adamantylcarboxylic acid (3j) under the present conditions.

A plausible catalytic cycle for the magnesiation of alkyl fluorides is shown in Scheme [5]. Initially, the catalyst precursor 1 is reduced by Mg powder to generate the X-type Al–Rh complex I, which cleaves the C(sp³)–F bond of alkyl fluoride 2 through cooperative bond activation across the polarized Al–Rh bond to give II.[8] In the case of tertiary alkyl fluorides, such as 2e, a stepwise mechanism through the transfer of the fluorine atom to the Lewis acidic Al center, followed by the formation of the alkyl–Rh bond, might be operative.[12] Finally, the Mg powder reduces II, resulting in the formation of the corresponding alkylmagnesium species III with concurrent regeneration of I, thereby closing the catalytic cycle. In the context of the Rh/P ligand system, Rh(-I)–Mg(II)X species might be formed and could be responsible for a cooperative C–F bond activation (for details, see the SI).[13] To gain insights into the proposed reaction mechanism, we conducted a hot-filtration test. The results supported the notion that the active species could be homogeneous (SI; Figure S6).[14]

In conclusion, we have developed a catalytic magnesiation of alkyl fluorides by using Mg powder.[15] Notably, the protocol can be applied to a range of alkyl fluorides, including secondary and tertiary ones.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

I.F. is grateful for the Research Fellowship of the Japan Society for the Promotion of Science (JSPS) for Young Scientists, the G-7 Scholarship Foundation, and the Iwadare Scholarship Foundation.

• References and Notes

• 1 Present address: I. Fujii, Division of Chemistry, Department of Materials Engineering Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka 560-8531, Japan.
• 2a Grignard V. C. R. Hebd. Seances Acad. Sci. 1900; 130: 1322
  Search in Google Scholar
• 2b Seyferth D. Organometallics 2009; 28: 1598
  CrossrefPubMedSearch in Google Scholar
• 3a Krasovskiy A, Knochel P. Angew. Chem. Int. Ed. 2004; 43: 3333
  CrossrefPubMedSearch in Google Scholar
• 3b Bao RL.-Y, Zhao R, Shi L. Chem. Commun. 2015; 51: 6884
  CrossrefPubMedSearch in Google Scholar
• 3c Ziegler DS, Wei B, Knochel P. Chem. Eur. J. 2019; 25: 2695
  CrossrefPubMedSearch in Google Scholar
• 4a Pietrasiak E, Lee E. Chem. Commun. 2022; 58: 2799
  CrossrefPubMedSearch in Google Scholar
• 4b Blanksby SJ, Ellison GB. Acc. Chem. Res. 2003; 36: 255
  CrossrefPubMedSearch in Google Scholar
• 4c O’Hagan D. Chem. Soc. Rev. 2008; 37: 308
  CrossrefPubMedSearch in Google Scholar
• 5a Pattison FL. M, Howell WC. J. Org. Chem. 1956; 21: 879
  CrossrefSearch in Google Scholar
• 5b Ashby EC, Yu SH, Beach RG. J. Am. Chem. Soc. 1970; 92: 433
  CrossrefSearch in Google Scholar
• 5c Bernstein J, Roth JS, Miller WT. Jr. J. Am. Chem. Soc. 1948; 70: 2310
  CrossrefSearch in Google Scholar
• 5d Ashby EC, Yu SH. J. Org. Chem. 1971; 36: 2123
  CrossrefSearch in Google Scholar
• 6a Rieke RD, Bales SE. J. Am. Chem. Soc. 1974; 96: 1775
  CrossrefSearch in Google Scholar
• 6b Rieke RD, Hudnall PM. J. Am. Chem. Soc. 1972; 94: 7178
  CrossrefSearch in Google Scholar
• 6c Klabunde KJ, Whetten A. J. Am. Chem. Soc. 1986; 108: 6529
  CrossrefSearch in Google Scholar
• 6d Bare WD, Andrews L. J. Am. Chem. Soc. 1998; 120: 7293
  CrossrefSearch in Google Scholar
• 7a Coates G, Ward BJ, Bakewell C, White AJ. P, Crimmin MR. Chem. Eur. J. 2018; 24: 16282
  CrossrefPubMedSearch in Google Scholar
• 7b Sheldon DJ, Parr JM, Crimmin MR. J. Am. Chem. Soc. 2023; 145: 10486
  CrossrefPubMedSearch in Google Scholar
• 7c Bonyhady SJ, Jones C, Nembenna S, Stasch A, Edwards AJ, McIntyre GJ. Chem. Eur. J. 2010; 16: 938
  CrossrefPubMedSearch in Google Scholar
• 8 Fujii I, Semba K, Li Q.-Z, Sakaki S, Nakao Y. J. Am. Chem. Soc. 2020; 142: 11647
  CrossrefPubMedSearch in Google Scholar
• 9 Ashby EC, Yu S. J. Organomet. Chem. 1971; 29: 339
  CrossrefSearch in Google Scholar
• 10a Basha A, Lipton M, Weinreb SM. Tetrahedron Lett. 1977; 18: 4171
  CrossrefSearch in Google Scholar
• 10b Nahm S, Weinreb SM. Tetrahedron Lett. 1981; 22: 3815
  CrossrefPubMedSearch in Google Scholar
• 11a Lu X. Top. Catal. 2005; 35: 73
  CrossrefSearch in Google Scholar
• 11b Hayashi T, Konishi M, Kobori Y, Kumada M, Higuchi T, Hirotsu K. J. Am. Chem. Soc. 1984; 106: 158
  CrossrefSearch in Google Scholar
• 12 Pitsch CE, Wang X. Chem. Commun. 2017; 53: 8196
  CrossrefPubMedSearch in Google Scholar
• 13a Bogdanović B, Leitner W, Six C, Wilczok U, Wittmann K. Angew. Chem. Int. Ed. 1997; 36: 502
  CrossrefSearch in Google Scholar
• 13b Seki R, Takaya H, Nakao Y. ChemRxiv 2023; DOI: preprint
  Crossref
• 14a Crabtree RH. Chem. Rev. 2012; 112: 1536
  CrossrefPubMedSearch in Google Scholar
• 14b Reay AJ, Fairlamb IJ. S. Chem. Commun. 2015; 51: 16289
  CrossrefPubMedSearch in Google Scholar
• 15 2-Methyl-4-phenylbutanoic Acid (3b); Typical Procedure In a glove box, a 4 mL vial equipped with a stirrer bar was charged with Mg powder (61 mg, 2.5 mmol, 5.0 equiv) and (3-fluorobutyl)benzene (2b; 76 mg, 0.50 mmol, 1.0 equiv). A suspension of catalyst 1a (20 mg, 13 μmol, 5.0 mol% of Rh) in THF (1.5 mL) was added to the vial, which was then capped with a PTFE sealing screw cap and removed from the glovebox. The mixture was stirred at 80 °C for 48 h, and the resulting mixture was then stirred under CO₂ at atmospheric pressure and r.t. for 2 h. 3 M aq HCl (1.5 mL) was added, and the resulting mixture was extracted with EtOAc (3 × 2.0 mL). All volatiles were removed in vacuo, and the residue was purified by MPLC [silica gel, hexane–EtOAc (70:30) + AcOH (10 vol%)] to give a colorless oil; yield: 65 mg (73%); R_f = 0.44 (hexane–EtOAc, 70:30 + 10% HOAc). ¹H NMR (400 MHz, CDCl₃): δ = 7.33–7.27 (m, 2 H), 7.23–7.14 (m, 3 H), 2.68 (t, J = 8.0 Hz, 2 H), 2.52 (sext, J = 6.9 Hz, 1 H), 2.14–1.99 (m, 1 H), 1.76 (ddt, J = 13.9, 8.4, 6.8 Hz, 1 H), 1.24 (d, J = 6.9 Hz, 3 H). ¹³C{¹H} NMR (101 MHz, CDCl₃): δ = 183.0, 141.6, 128.6, 128.5, 126.1, 38.9, 35.3, 33.5, 17.1.

Corresponding Author

Publication History

Received: 05 August 2023

Accepted after revision: 28 December 2023

Accepted Manuscript online:
28 December 2023

Article published online:
29 January 2024

© 2024. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References and Notes

• 1 Present address: I. Fujii, Division of Chemistry, Department of Materials Engineering Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka 560-8531, Japan.
• 2a Grignard V. C. R. Hebd. Seances Acad. Sci. 1900; 130: 1322
  Search in Google Scholar
• 2b Seyferth D. Organometallics 2009; 28: 1598
  CrossrefPubMedSearch in Google Scholar
• 3a Krasovskiy A, Knochel P. Angew. Chem. Int. Ed. 2004; 43: 3333
  CrossrefPubMedSearch in Google Scholar
• 3b Bao RL.-Y, Zhao R, Shi L. Chem. Commun. 2015; 51: 6884
  CrossrefPubMedSearch in Google Scholar
• 3c Ziegler DS, Wei B, Knochel P. Chem. Eur. J. 2019; 25: 2695
  CrossrefPubMedSearch in Google Scholar
• 4a Pietrasiak E, Lee E. Chem. Commun. 2022; 58: 2799
  CrossrefPubMedSearch in Google Scholar
• 4b Blanksby SJ, Ellison GB. Acc. Chem. Res. 2003; 36: 255
  CrossrefPubMedSearch in Google Scholar
• 4c O’Hagan D. Chem. Soc. Rev. 2008; 37: 308
  CrossrefPubMedSearch in Google Scholar
• 5a Pattison FL. M, Howell WC. J. Org. Chem. 1956; 21: 879
  CrossrefSearch in Google Scholar
• 5b Ashby EC, Yu SH, Beach RG. J. Am. Chem. Soc. 1970; 92: 433
  CrossrefSearch in Google Scholar
• 5c Bernstein J, Roth JS, Miller WT. Jr. J. Am. Chem. Soc. 1948; 70: 2310
  CrossrefSearch in Google Scholar
• 5d Ashby EC, Yu SH. J. Org. Chem. 1971; 36: 2123
  CrossrefSearch in Google Scholar
• 6a Rieke RD, Bales SE. J. Am. Chem. Soc. 1974; 96: 1775
  CrossrefSearch in Google Scholar
• 6b Rieke RD, Hudnall PM. J. Am. Chem. Soc. 1972; 94: 7178
  CrossrefSearch in Google Scholar
• 6c Klabunde KJ, Whetten A. J. Am. Chem. Soc. 1986; 108: 6529
  CrossrefSearch in Google Scholar
• 6d Bare WD, Andrews L. J. Am. Chem. Soc. 1998; 120: 7293
  CrossrefSearch in Google Scholar
• 7a Coates G, Ward BJ, Bakewell C, White AJ. P, Crimmin MR. Chem. Eur. J. 2018; 24: 16282
  CrossrefPubMedSearch in Google Scholar
• 7b Sheldon DJ, Parr JM, Crimmin MR. J. Am. Chem. Soc. 2023; 145: 10486
  CrossrefPubMedSearch in Google Scholar
• 7c Bonyhady SJ, Jones C, Nembenna S, Stasch A, Edwards AJ, McIntyre GJ. Chem. Eur. J. 2010; 16: 938
  CrossrefPubMedSearch in Google Scholar
• 8 Fujii I, Semba K, Li Q.-Z, Sakaki S, Nakao Y. J. Am. Chem. Soc. 2020; 142: 11647
  CrossrefPubMedSearch in Google Scholar
• 9 Ashby EC, Yu S. J. Organomet. Chem. 1971; 29: 339
  CrossrefSearch in Google Scholar
• 10a Basha A, Lipton M, Weinreb SM. Tetrahedron Lett. 1977; 18: 4171
  CrossrefSearch in Google Scholar
• 10b Nahm S, Weinreb SM. Tetrahedron Lett. 1981; 22: 3815
  CrossrefPubMedSearch in Google Scholar
• 11a Lu X. Top. Catal. 2005; 35: 73
  CrossrefSearch in Google Scholar
• 11b Hayashi T, Konishi M, Kobori Y, Kumada M, Higuchi T, Hirotsu K. J. Am. Chem. Soc. 1984; 106: 158
  CrossrefSearch in Google Scholar
• 12 Pitsch CE, Wang X. Chem. Commun. 2017; 53: 8196
  CrossrefPubMedSearch in Google Scholar
• 13a Bogdanović B, Leitner W, Six C, Wilczok U, Wittmann K. Angew. Chem. Int. Ed. 1997; 36: 502
  CrossrefSearch in Google Scholar
• 13b Seki R, Takaya H, Nakao Y. ChemRxiv 2023; DOI: preprint
  Crossref
• 14a Crabtree RH. Chem. Rev. 2012; 112: 1536
  CrossrefPubMedSearch in Google Scholar
• 14b Reay AJ, Fairlamb IJ. S. Chem. Commun. 2015; 51: 16289
  CrossrefPubMedSearch in Google Scholar
• 15 2-Methyl-4-phenylbutanoic Acid (3b); Typical Procedure In a glove box, a 4 mL vial equipped with a stirrer bar was charged with Mg powder (61 mg, 2.5 mmol, 5.0 equiv) and (3-fluorobutyl)benzene (2b; 76 mg, 0.50 mmol, 1.0 equiv). A suspension of catalyst 1a (20 mg, 13 μmol, 5.0 mol% of Rh) in THF (1.5 mL) was added to the vial, which was then capped with a PTFE sealing screw cap and removed from the glovebox. The mixture was stirred at 80 °C for 48 h, and the resulting mixture was then stirred under CO₂ at atmospheric pressure and r.t. for 2 h. 3 M aq HCl (1.5 mL) was added, and the resulting mixture was extracted with EtOAc (3 × 2.0 mL). All volatiles were removed in vacuo, and the residue was purified by MPLC [silica gel, hexane–EtOAc (70:30) + AcOH (10 vol%)] to give a colorless oil; yield: 65 mg (73%); R_f = 0.44 (hexane–EtOAc, 70:30 + 10% HOAc). ¹H NMR (400 MHz, CDCl₃): δ = 7.33–7.27 (m, 2 H), 7.23–7.14 (m, 3 H), 2.68 (t, J = 8.0 Hz, 2 H), 2.52 (sext, J = 6.9 Hz, 1 H), 2.14–1.99 (m, 1 H), 1.76 (ddt, J = 13.9, 8.4, 6.8 Hz, 1 H), 1.24 (d, J = 6.9 Hz, 3 H). ¹³C{¹H} NMR (101 MHz, CDCl₃): δ = 183.0, 141.6, 128.6, 128.5, 126.1, 38.9, 35.3, 33.5, 17.1.

• Supplementary Material